the interactive effects of soil transplant into colder regions and cropping on soil microbiology and...

The interactive effects of soil transplant intocolder regions and cropping on soil microbiologyand biogeochemistry

Shanshan Liu,1 Feng Wang,2,3 Kai Xue,4 Bo Sun,2

Yuguang Zhang,5 Zhili He,4 Joy D. Van Nostrand,4

Jizhong Zhou1,4,6 and Yunfeng Yang1*1State Key Joint Laboratory of Environment Simulationand Pollution Control, School of Environment, TsinghuaUniversity, Beijing, China.2State Key Laboratory of Soil and SustainableAgriculture, Institute of Soil Science, Chinese Academyof Sciences, Nanjing, China.3University of Chinese Academy of Sciences, Beijing,China.4Institute for Environmental Genomics, DepartmentMicrobiology and Plant Science, University ofOklahoma, Norman, OK, USA.5Institute of Forestry Ecology, Environment andProtection, Key Laboratory of Forest Ecology andEnvironment of State Forestry Administration, ChineseAcademy of Forestry, Beijing, China.6Earth Sciences Division, Lawrence Berkeley NationalLaboratory, Berkeley, CA, USA.

Summary

Soil transplant into warmer regions has been shownto alter soil microbiology. In contrast, little is knownabout the effects of soil transplant into colder regions,albeit that climate cooling has solicited attentionin recent years. To address this question, we trans-planted bare fallow soil over large transects fromsouthern China (subtropical climate zone) to central(warm temperate climate zone) and northern China(cold temperate climate zone). After an adaptationperiod of 4 years, soil nitrogen components, microbialbiomass and community structures were altered.However, the effects of soil transplant on microbialcommunities were dampened by maize cropping,unveiling a negative interaction between croppingand transplant. Further statistical analyses withCanonical correspondence analysis and Mantel testsunveiled annual average temperature, relative humid-

ity, aboveground biomass, soil pH and NH4+-N con-

tent as environmental attributes closely correlatedwith microbial functional structures. In addition, aver-age abundances of amoA-AOA (ammonia-oxidizingarchaea) and amoA-AOB (ammonia-oxidizing bacte-ria) genes were significantly (P < 0.05) correlated withsoil nitrification capacity, hence both AOA and AOBcontributed to the soil functional process of nitrifica-tion. These results suggested that the soil nitrogencycle was intimately linked with microbial communitystructure, and both were subjected to disturbance bysoil transplant to colder regions and plant cropping.

Introduction

Global change induces shifts in microbial community com-position, which in turn alters microbially mediated green-house gas emissions in nature (Hansen et al., 2006;Walther, 2010). By using a strategy of soil transplant intowarmer regions to simulate climate warming, it has beenshown that soil microbial community structure and com-munity functions were altered (Vanhala et al., 2011),which was consistent with a number of studies showingthat climate warming alters microbial communities(Petchey et al., 1999; Rinnan et al., 2007; Zhou et al.,2012). It is also of interest to investigate the effects of soiltransplant into colder regions, which may contribute to ourunderstanding of climate cooling, since the impact ofclimate cooling on an ecosystem can be at a scale com-parable to that of climate warming (McAnena et al., 2013).Although long-term cooling worldwide may have a lowprobability of occurrence, it could occur locally or tempo-rally, and may have large and devastating consequences(Alley et al., 2003). For instance, the cooling of seasurface temperatures in the centre of the North PacificOcean in 1977 has resulted in serious consequences formarine ecosystems (Hare and Mantua, 2000).

In particular, we are interested in understandingwhether and how the composition and functional poten-tials of the microbial community are linked to communitymetabolism. Although a growing body of studies has dem-onstrated that microbial community composition plays anessential role in biogeochemical cycles (Falkowski et al.,2008), the nature and quantification of this linkage is

Received 17 June, 2013; revised 8 January, 2014; accepted 8January, 2014. *For correspondence. E-mail [email protected]; Tel. (+86) 010 62784692; Fax (+86) 010 62785687.

bs_bs_banner

Environmental Microbiology (2014) doi:10.1111/1462-2920.12398

© 2014 Society for Applied Microbiology and John Wiley & Sons Ltd

mailto:[email protected]

mailto:[email protected]

neither consistent nor tractable. This knowledge gap hasresulted in the disputable assumption of global climaticmodels that microbial community composition is irrelevantto global climate change (Reed and Martiny, 2007).

In this study, we specifically addressed the followingquestions: (i) whether and how soil attributes, microbialbiomass and community structures responded to soiltransplant into colder regions, (ii) whether and how agri-cultural cropping modified the effect of soil transplantand (iii) whether changes in microbial communities werelinked to changes in environmental attributes related tothe disturbances of soil transplant and cropping. To thisend, we transplanted soil in 2005 between three long-standing agricultural stations – from Yingtan, JiangxiProvince, southern China (site S) to Fengqiu, HenanProvince of central China (site SC) and to Hailun,Heilongjiang Province of northern China (site SN), expos-ing soil from the subtropical climate zone to warm andcold temperate climate zones respectively. The soil typestudied was ferric acrisol in the FAO (the Food and Agri-culture Organization of the United Nations) classificationsystem (red soil), which is widely distributed across Asia,Africa, North America and South America.

Results

Effects of soil transplant on soil attributes

Although climate regimes differed substantially betweenthe transplant sites and the control site, most of the soilbiogeochemical attributes, including organic matter, totalnitrogen (TN), total phosphorus (TP) and total potassium(TK), remained unaltered 4 years after soil transplant(Table 1). However, NH4

+-N content decreased from2.98 to 2.46–2.47 mg kg−1 at sites SC and SN, alkali-hydrolysable nitrogen (AN) decreased from 52.81 mg kg−1

to 49.51 mg kg−1 at site SC and 39.61 mg kg−1 at site SNbut NO3

−-N content increased from 12.06 to 35.07 mg kg−1

at site SC and decreased to 7.75 mg kg−1 at site SN.These results indicated that soil nitrogen componentswere more sensitive than other elemental nutrients inresponse to soil transplant. The effects of nitrogen com-ponents were further explored by correlation analyses.The results showed that annual average temperature andrainfall were positively correlated with TN (r = 0.47–0.48,P < 0.05) and AN (r = 0.67–0.73, P < 0.002), and soilmoisture was correlated with AN (r = 0.52, P = 0.03).Thus, these climatic attributes imposed an effect on soilnitrogen, which was consistent with previous reports(Shaw and Harte, 2001; Rowe et al., 2012).

Effects of soil transplant on soil microbial communities

The functional composition of soil microbial communitieswas analysed using a high-throughput metagenomics tool

named GeoChip. Microbial functional diversity, deter-mined by Simpson’s index, decreased from 2094.5 at siteS to 1811.0 at site SC and to 1685.0 at site SN (Support-ing information Fig. S1). In addition, site S, SC and SNsamples were well separated by detrended correspond-ence analysis (DCA) (Fig. 1), which was confirmed by thedissimilarity test of Adonis, showing that the differenceswere significant (P < 0.001) (Table 2).

The results of phospholipid fatty acid (PLFA) analysisshowed that soil at site SC had a lower total microbialbiomass (11.79 nmol g−1 dry soil) and fungal biomass(0.71 nmol g−1 dry soil) than soil at site S (21.28 nmol totalmicrobial biomass/g dry soil and 1.77 nmol fungalbiomass/g dry soil), while bacterial biomass remainedunchanged between these two sites. In contrast, the totalmicrobial, fungal and bacterial biomass remainedunchanged at site SN (Supporting information Fig. S2).The ratios of bacteria to fungi were increased at both siteSC (2.2 fold, P = 0.004) and site SN (1.6 fold, P = 0.06)(Fig. 2A). These results revealed the differential sensitivityof bacteria and fungi to cooler and drier climate regimes.

To examine whether environmental attributes in-fluenced bacterial and fungal biomass, we analysedPearson correlations between bacterial/fungal ratios andenvironmental attributes. As shown in Fig. 2B, the ratiosof bacteria to fungi were negatively and positively corre-lated with soil moisture (r = −0.61, P = 0.01) and pH(r = 0.52, P = 0.02) respectively. These two attributes

Table 1. Environmental attributes of S, SC and SN samples.

Sa SC SN

Climate attributesAnnual average Tb (°C) 18.04a 14.05 1.46b

Rainfall (mm) 132.41a 128.8 60.4b

Relative humidity (%) 77.80b 80.05 85.13a

Soil physical-chemical attributesSoil T (°C) 29.25a 30.07a 27.37pH 5.97a 6.54a 5.80a

Moisture (%) 21.62a 9.59 7.45b

Organic matter (g kg−1) 9.92a 10.05a 8.91a

TN (g kg−1) 0.64a 0.69a 0.59a

TP (g kg−1) 0.44a 0.49a 0.34TK (g kg−1) 9.86a 13.04a 9.95a

AN (mg kg−1) 52.81a 49.51a 39.61AP (mg kg−1) 21.06a 17.14a 12.94a

AK (mg kg−1) 195a 227.5a 186.67a

NH4+-N (mg kg−1) 2.98a 2.46 2.47

NO3−-N (mg kg−1) 12.06 35.07a 7.75b

Soil ecoprocessesNitrification capacity (%) 9.05a 9.09a 9.78a

Soil CO2 efflux (umolCO2/(m2 s)−1) 1.02a 1.79a 1.38a

NH4+-N, ammonium; NO3

−-N, nitrate; T, temperature.a. S: soil samples in southern China (Yintan); SC: soil samplestransplanted from southern China to central China (Fengqiu); SN: soilsamples transplanted from southern China to northern China(Hailun).b. Attributes with significant differences are shown in bold font.

2 S. Liu et al.

© 2014 Society for Applied Microbiology and John Wiley & Sons Ltd, Environmental Microbiology

were both linked to the relative abundance of bacterialand fungal communities; however, their effects were felt inopposite directions.

Effects of soil transplant on selected functional genes

Next, we focused on functional genes present only at bothtransplant sites, which might be important for microbialadaptation to new environments. This gene groupincluded 89 genes, many of which were derived fromProteobacteria and Actinobacteria prevalent in soil(Roesch et al., 2007). In contrast, no genes were derivedfrom fungi, and only three genes were derived fromarchaea. Most of these genes are involved in organicremediation related to the degradation of complex carbonsubstrates (34.83%), carbon cycling (21.35%), nitrogencycling (7.87%), antibiotic resistance (8.99%) and metalresistance (16.85%) (Supporting information Table S1).Compared with the overall percentages of gene cate-gories in the control soil, genes related to organic reme-diation and the carbon cycle were overrepresented(increased by 7.38% and 1.86% respectively), implyingthat carbon cycling mediated by microbes was required atthe transplant sites. In contrast, nitrogen cycling andmetal resistance genes were underrepresented atthe transplant sites (decreased by 3.82% and 5.98%respectively).

A total of 582 genes were exclusively detected at site S(Supporting information Table S2). Among these genes,

84 were carbon degradation genes. Interestingly, 15 ofthese carbon degradation genes were derived from fungisuch as Eurotiomycetes and Agaricomycetes, suggestingthat the carbon degradation potential of these fungalspecies was reduced. There might be a shift in the soilcarbon use profile from recalcitrant carbon to labilecarbon because considerably more genes involved indegrading recalcitrant carbon (cellulose, chitin and lignin)Fig. 1. The effect of soil transplant into colder regions on microbial

community functional structure revealed by detrendedcorrespondence analysis in bare fallow samples. S: soil samples insouthern China (Yintan); SC: soil samples transplanted fromsouthern China to central China (Fengqiu); SN: soil samplestransplanted from southern China to northern China (Hailun).

Table 2. The statistical test of Adonisa to analyse differences in soilmicrobial community compositions measured by GeoChip.

Method

GeoChip

Statistic R2 P-value

S vs SC 0.786 0.001S vs SN 0.715 0.001SC vs SN 0.703 0.001

a. Adonis: non-parametric multivariate analysis of variance with theAdonis function.

Fig. 2. A. Changes in bacterial/fungal ratios in bare fallow samples.Significant differences between S and SC or between S and SNwere analysed by two-tailed t-tests with two-sample equal varianceand indicated by ‘*’ for P < 0.1 and ‘***’ for P < 0.01 respectively.Symbols as in Fig. 1. B. Pearson correlations between ratios ofbacteria to fungi and pH and soil moisture. The ratios werepositively correlated with pH (P = 0.02) and negatively correlatedwith soil moisture (P = 0.01). Filled and open circles represent soilpH and moisture respectively.

Effects of northward soil transplant and cropping 3


disappeared after soil transplant when compared withthose involved in degrading labile carbon (starch andhemicellulose). For carbon fixation, 14.9% of detectedCODH genes disappeared after soil transplant, which washigher than the disappearance of genes pcc (6.3%) andrubisco (12.3%).

The interactive effect of soil transplant andmaize cropping

GeoChip data from samples under the dual factors of soiltransplant and maize cropping (SCm and SNm) was sig-nificantly (P < 0.04) different from that of a single factor(Supporting information Table S3), suggesting that maizecropping interacted with the effect of soil transplant onmicrobial communities. This was verified by DCA (Sup-porting information Fig. S3) and hierarchical clusteringanalysis (Fig. 3), which showed that samples weregrouped by their respective treatment(s).

To explore the possibility of interactions betweenclimate cooling and maize cropping, we calculated theadditive effect of soil transplant and maize cropping onmicrobial alpha-diversities with the assumption that theinteractive effect was zero. The observed effect was con-siderably lower than the additive effect (Supporting infor-mation Fig. S4), indicating a negative interaction betweensoil transplant and maize cropping.

Relationships between microbial functional genes andenvironmental attributes

Canonical correspondence analysis was performed toreveal the relationships between microbial communitycomposition and environmental attributes. In this analysis,the environmental attributes with and without maizecropping were all included (Table 1 and Supporting

information Table S4). Six environmental attributes (an-nual average temperature, pH, relative humidity, AN,NH4

+-N and aboveground biomass) were selected basedon their variance inflation factors (VIFs) (See Methodsand Materials for details). The results indicated thatclimate conditions (annual average temperature and rela-tive humidity), soil conditions (pH, NH4

+-N and AN) andvegetation (aboveground biomass) were major factorsinfluencing microbial community structure (Fig. 4). Theseresults were further verified by Mantel tests (Table 3).

We performed correlation analyses to examine theinfluence of environmental attributes on individual func-tional genes. A total of 134 genes were significantly(P < 0.01) correlated with annual average temperature

Fig. 3. Hierarchical clustering analysis of microbial communities for all soil samples in triplicate based on GeoChip data. Results weregenerated in R 2.12.2 with Vegan package. Symbols as in Fig. 1; m: maize cropping.

Fig. 4. CCA of the GeoChip data (symbols) and environmentalattributes (arrows). The percentage of variation explained by eachaxis is shown. The CCA model is significant (P = 0.005). Symbolsas in Fig. 3; Annual average T: annual average temperature; AN:alkali-hydrolysable nitrogen; NH4

+-N: ammonium.

4 S. Liu et al.


(Supporting information Table S5). The correlations werepredominantly positive, suggesting that lower tempera-ture decreased gene abundance and hence microbialfunctional potentials at our study sites. There were seve-ral ribulose-1, 5-bisphosphate carboxylase/oxygenase(rubisco) genes for the Calvin cycle and carbon monoxidedehydrogenase (CODH) genes for the reductive acetyl-CoA pathway correlated with temperature, suggestingthat these carbon fixation pathways were sensitive todecreased temperature. In contrast, no propionyl-CoAcarboxylase (pcc) gene for the 3-hydroxypropionate cyclewas correlated with temperature, despite the presence ofa large number of pcc probes on GeoChip. A number ofgenes involved in the methane cycle and labile carbondegradation were also identified, for instance, pmoA formethane oxidation, which has been studied to be tem-perature independent (King and Adamsen, 1992);cellobiase and endoglucanase for cellulose degradation,which were found to be correlated with temperature(Yergeau et al., 2007a). In contrast, there were few genesinvolved in recalcitrant carbon degradation (chitin andlignin) found to be correlated with temperature in thisstudy. For the nitrogen cycle, only 19 genes were corre-lated with temperature, and most were involved in nitro-gen fixation (nifH) and denitrification (nirK and nirS),which was consistent with recent studies that found thatnifH and nir genes were sensitive to temperature(Yergeau et al., 2007a; Jung et al., 2011).

A total of 165 genes were correlated with pH (|r| = 0.57–0.92, P < 0.01) (Supporting information Table S6). Inter-estingly, a majority of these genes were negativelycorrelated with pH, implying that the microbes hostingthese genes preferred the acidic habitats of the donor site

(site S). Similar to recent observations that pH had astrong influence on the copA genetic structure (Lejonet al., 2007), we found a number of copA genes to be pHsensitive, most of which were derived from Prote-obacteria. Three nirK genes involved in denitrificationwere correlated with pH. In contrast, only one nirS genewas correlated with pH, although there are similarnumbers of nirK and nirS gene probes on GeoChip.This observation was consistent with recent findings thatnirK, but not nirS, was correlated with pH (Dandie et al.,2011).

A striking number of genes, 1163 (14.2% of totaldetected genes), were correlated with NH4

+-N (|r| = 0.56–0.99, P < 0.01), suggesting that there were stronglinkages between microbial communities and NH4

+-Nconcentrations. Based on the gyrB gene, a usefulmarker of microbial phylogenetic distribution (Hirschet al., 2010; Peeters and Willems, 2011), a number ofShewanella, Pseudomonas and Rhodobacter-like speciesof Proteobacteria were significantly correlated withNH4

+-N (data not shown). This was consistent with theirimportant roles in the nitrogen cycle (Chèneby et al.,2000; Dandie et al., 2007; Gao et al., 2009).

Linking microbial genes to soil functional processes

Soil nitrification is mainly a microbially mediated soil func-tional process. In particular, ammonia-oxidizing archaea(AOA) and ammonia-oxidizing bacteria (AOB) playedimportant roles in controlling the rates of nitrification(Zhang et al., 2011). In this study, the average abun-dances of both amoA-AOA (r = 0.48, P = 0.05) and amoA-AOB (r = 0.65, P = 0.004) were significantly correlatedwith nitrification capacity, with a higher correlationobtained for amoA-AOB (Fig. 5). However, only four indi-vidual amoA genes were detected in all samples, includ-ing two amoA-AOA derived from Crenarchaeota andamoA-AOB derived from unknown bacteria, which playedimportant roles in controlling nitrification capacity.

We proceeded to examine whether there was a linkagebetween microbial carbon degradation genes and in situsoil CO2 efflux. However, no significant correlation wasdetected between the average abundance of microbialcarbon degradation genes and soil CO2 efflux (data notshown), suggesting that other factors played a vital role indetermining soil CO2 efflux.

Discussion

The use of soil transplant as a proxy to study the effectsof climate change has been successfully demonstrated inboth plant biology and microbiology (Breeuwer et al.,2010; De Frenne et al., 2011; Vanhala et al., 2011). In thisstudy, we showed that 4 years of exposure to colder

Table 3. Correlations between all detected genes and environmentalattributes by Mantel test.

Statistic r P-value

Annual average T 0.156 0.062Rainfall 0.046 0.301Relative humidity 0.144 0.048*a

Soil T 0.0097 0.389pH 0.358 0.012*Moisture 0.079 0.19Organic matter 0.033 0.334TN 0.131 0.119TP −0.074 0.622TK 0.068 0.182AN 0.216 0.036*AP −0.050 0.611AK −0.087 0.677NH4

+-N 0.734 0.001**NO3

−-N 0.148 0.1Grain weight 0.092 0.219Aboveground biomass 0.230 0.067

T, temperature; NH4+-N, ammonium; NO3

−-N, nitrate.a. Asterisk indicates significant differences. ‘*’, P < 0.05; ‘**’,P < 0.01.



climates were sufficient to alter soil nitrogen componentsand microbial communities. The effects of soil transplantwere negatively affected by maize cropping. Changes inmicrobial communities were explainable by the selectedenvironmental attributes, and significant linkages werenotable between soil nitrification and related microbialfunctional groups.

The climatic effect on microbial community compositionis under heated debate. It has been shown that soil micro-bial diversity decreased when latitude increased and waspositively correlated with air temperature (Yergeau et al.,2007b; Bryant et al., 2008). Similarly, a significant latitu-dinal diversity gradient, with a strong correlation with tem-perature, was detected in planktonic marine bacteria(Fuhrman et al., 2008). In contrast, there were large-scalestudies indicating that bacterial diversity was not corre-lated with temperature or latitude (Fierer and Jackson,2006; Chu et al., 2010; Mi et al., 2012). Our observationthat the functional diversity of microbial communitydetected by GeoChip decreased at higher latitudes sup-ported studies showing that soil microbial diversitydecreased with the increase of latitude, albeit that theeffects of other climate variables (e.g. precipitation and airnitrogen deposition) could not be excluded because of thelimited number and size of our study sites. The effects ofsoil transplant on microbial communities could be attrib-uted to lesser primary productivity, poorer litter quality andcolder climate at high latitudes (Ma et al., 2004), leadingto elevated environmental harshness for microbial com-munities. However, primary productivity and litter couldnot be the major causes in this study as the study sites

were bare fallow (Supporting information Fig. S1). Climateconditions at the colder site might be the primary cause ofchanges in microbial community; however, it remainedunclear whether changes in microbial communities weremediated by concurrent changes in soil attributes.

By exposing soil to new climate regimes in the trans-plant experiments, soil inhabitants were pressured toacclimate or adapt to their new environments. The home-site advantage has been well observed in plants(Montalvo and Ellstrand, 2001; Savolainen et al., 2007)and also provided an explanation for the observation inour experiments that microbial biomass was reduced bytransplant in general. Alternatively, the decrease in micro-bial biomass might be caused by the transplant proce-dure. Soil mixing and transportation during transplantmight have an adverse effect on microbial communities.To minimize this possibility, we mocked soil transplant forthe control site and waited for 4 years so that the trans-plant effect could subside. The results of bacterial/fungalbiomass ratios suggested that the alteration in microbialbiomass could be linked to environmental attributes, asthere were significant correlations with soil moisture andpH (Fig. 2B). Notably, it has been well established that thedry–wet climatic gradient negatively altered the ratios ofbacteria to fungi (Frey et al., 1999), which was in line withour finding.

Soil transplant caused alterations in microbial biomassand functional community composition. In contrast, onlysoil nitrogen components, but not those of carbon, phos-phorus or potassium, showed notable sensitivity to soiltransplant. These results were consistent with recentfindings that soil microbial parameters were sensitiveindicators of the change in soil conditions (Lagomarsinoet al., 2009; Li et al., 2009). These soil microbial param-eters could replace soil geochemical indicators sincecarbon, phosphorus and potassium varied slowly overtime. Soil microbial status, including microbial biomass(García-Orenes et al., 2010), enzyme activity (García-Ruiz et al., 2008) and community diversity (Sharmaet al., 2011), has been tested as an indicator of soil con-ditions. Here, we show that functional composition ofmicrobial communities, as detected by GeoChip, canpotentially be used to indicate the changes in soilconditions.

Although changes in soil pH were not significant, ourcorrelation analyses indicated that it was an important soilattribute in shaping microbial communities. Consistently,soil pH has been shown to impose a strong influence onmicrobial community abundance and composition (Fiererand Jackson, 2006; Nilsson et al., 2007), either directly orindirectly through concurrent environmental attributessuch as nutrient availability (e.g. the shift between NH3

and NH4+). At the pH interval of 4–7, it was shown that

higher soil pH was correlated with higher bacterial diver-

Fig. 5. Pearson correlations between nitrification capacity and theaverage abundance of amoA-AOA and amoA-AOB. Filled and opencircles represent the average abundance of amoA-AOA andamoA-AOB respectively.

6 S. Liu et al.


sity (Lauber et al., 2009; Rousk et al., 2010), making it agood predictor of bacterial community composition. Incontrast, fungi tolerated wider pH ranges than bacteria,and their diversity was only weakly linked to pH (Nevarezet al., 2009; Rousk et al., 2010). The differential sensitivityof bacteria and fungi to pH was also demonstrated in ourfinding that bacterial/fungal biomass ratios were positivelycorrelated with soil pH (Fig. 2B).

Maize cropping dampened the effect of soil trans-plant on microbial communities (Supporting informationFig. S4), suggesting that it played a major role in shapingmicrobial communities. Cropping can influence microbialcommunities through a number of intertwined mecha-nisms. As a result of the increased aboveground litterinput to microbes, cropping has been shown to signifi-cantly increase microbial diversity (Zhang et al., 2005).Meanwhile, cropping directly impacts the community com-position of root-associated organisms or indirectly impactsthe community composition of root-associated organismsby root exudates (Wardle et al., 2004). Therefore, crop-ping could interact with climate change-induced effects byenabling soil microbial communities to be more resistantto climate change (Potthoff et al., 2006).

amoA-AOA and amoA-AOB were correlated with soilnitrification capacity in our experiments, raising the pos-sibility that DNA abundance can be used to indicate ratesof soil ecoprocesses such as greenhouse gas emission.However, it has been commonly assumed that abund-ance of microbial community DNA measures the meta-bolic potential, and only abundance of messenger RNA(mRNA) or protein represents functional activity. Unfortu-nately, the detection of soil mRNA from field samplesimposes a number of challenges including interferencefrom ribosomal RNA (rRNA) and transfer RNA (tRNA),strict transport and storage requirements, rapid turnoverand severe instability (Sessitsch et al., 2002; Mettel et al.,2010), rendering the profiling and quantification of mRNAvery difficult, if not intractable, in many samples. In con-trast, DNA profiling and quantification are substantiallyreliable compared with mRNA. We argue that changes inDNA can be used to estimate functional activity becauseit indicates that the microbial population is active acrossthe samples, thus providing a good alternative to short-lived RNAs in estimating functional activity. In support ofthis viewpoint, DNA abundance of nirS–nosZ genes wasrecently shown to correlate with N2O emission in soils,suggesting that abundance of these functional genescould serve as a proxy for greenhouse gas (N2O) emis-sions (Morales et al., 2010). Similarly, other studies alsodemonstrated that DNA abundance could be used forassessing the activities of microbial communities (VanNostrand et al., 2011; Liang et al., 2012).

Recognizing the effects of environmental disturbanceon soil microbial community is crucial for microbial

resource conservation. Our results showed that microbialcommunities were significantly altered by soil transplantinto colder regions, but the effects of soil transplant weredampened by maize cropping. Their interactive impactsreshaped soil microbial communities and played animportant role in dictating the microbial response. Ourresults illustrate complex changes in terrestrial microbialecosystems under different climate change scenarios andthus highlight the need to further understand how micro-bial communities will potentially be affected and theirfeedbacks to climate change.

Experimental procedures

Site description and sampling

The transplant experiments were carried out in agriculturalexperimental stations of the Chinese Academy of Sciences atthree sites, from Yingtan, Jiangxi Province of southern Chinato Fengqiu, Henan Province of central China and Hailun,Heilongjiang Province of northern China, across different cli-matic regimes from subtropical to warm temperate and coldtemperate climate respectively. Soil at site S was character-ized as red soil derived from quaternary red clay, which isrepresentative of about 13.39% of soil in China. At each site,triplicate soil plots were established with and without maizecropping. In October 2005, soil from site S (1.2 × 1.4 × 1.0 msoil) was removed in five layers with a depth of 0.2 m eachlayer. Soil from each layer was mixed, transported to sites SCand SN and placed into soil holes of the same size accordingto the original sequence. The plots at site S were mocktransplanted but remained in place to serve as controls. Thetransplanted and control plots were separated from the sur-rounding environment by 20 cm brick walls. The plots werepaved underneath with quartz sand. The inner walls werecovered with waterproof membrane to prevent infiltration.Moreover, the plots were built to be 30 cm above the groundto prevent invasion by surrounding soil. Maize was planted in2006 at all three sites. Neither fertilization nor irrigation wasmanaged. Maize was harvested and sowed once a year. Thesamples with maize cropping from these three sites weredesignated as Sm, SCm and SNm. For bare fallow plots, nomanagement was undertaken except for manual de-weeding,whenever weeds were spotted, to prevent plant biomass orlitter input.

Soils from the three study sites were sampled in August–September, 2009. Ten soil cores were composited fromsurface soil (0–20 cm) within each plot and sealed in a poly-thene wrapper, then stored on ice and transported to thelaboratory. Any visible living plant material (e.g. roots) wasmanually removed from the composited soil in the lab. Thesoil was then divided into two subsamples and stored ateither 4°C for soil attribute measurements or −80°C forPLFAs and GeoChip analysis.

Soil geochemical attributes, vegetation andfunctional processes

All of the measurements of soil geochemical attributes werecompleted within 2 months after sampling. Soil moisture



content was measured gravimetrically with 10 g fresh soil byincubating in an oven at 105°C for 12 h. To determine nitrate(NO3

−-N) and ammonium (NH4+-N) concentrations, 20 g

fresh soil was suspended in 100 ml of 1 mol L−1 KCL for 1 hwith gentle shaking. NO3

−-N and NH4+-N concentrations

were measured in the suspension by Auto Analyser 3(Bran + Luebbe GmbH, Germany). Subsamples of soil for theanalysis of other geochemical attributes were air dried andkept at room temperature. Soil pH was measured usinga glass electrode (Mettler Toledo Instruments, Shanghai,China) with a ratio of 1:2.5 of soil (10 g, sieved to 1 mmmesh) to water. Soil organic matter was determined bydichromate oxidation with external heat and titration withferrous ammonium sulphate (Mebius, 1960). Soil TN wasdetermined by semimicro-Kjeldahl digestion (Page, 1982).Soil AN was measured using the Illinois Soil Nitrogen Testdiffusion method (Khan et al., 2001). Soil TP was determinedcolorimetrically using the molybdate method (Murphy andRiley, 1962). Soil available phosphorus (AP) was determinedcolorimetrically based on the Olsen method (Olsen et al.,1954). Soil TK was extracted by incubation with sodiumhydroxide, and soil available potassium (AK) was extractedby incubation with 1.0 mol L−1 ammonium acetate for 0.5 hfollowed by filtration. The concentrations in the extracts weredetermined by flame photometry (CANY Precision Instru-ment, Shanghai, China). Climate attributes, including annualaverage temperature, rainfall and relative humidity wereobtained from the experimental stations’ meteorologicalobservation data.

When sampling, all the crops in one plot were harvestedfirst, after which stalks were selected and immediately driedin the oven until the weight remained constant. Based on thewater content in the stalks, aboveground biomass and grainweight were calculated for each plot.

Soil temperature and CO2 efflux at 0–5 cm were measuredusing a soil respiration chamber (LI-6400-09; LiCor, Lincoln,NE, USA) connected to a portable photosynthesis system(LI-6400) at the time of sampling. To measure soil nitrificationpotential, soil (70 g; oven dry weight equivalent) was placedin a pre-weighed beaker, then 31.5 mg (NH4)2SO4 and addi-tional water were added to adjust the final soil moisturecontent to be 70% of water holding capacity. The beakerswere capped with perforated Parafilm, mixed well and incu-bated at 25°C for 14 days. Samples were monitored to main-tain constant moisture content. At days 0, 2, 3, 7, 10 and 13,the concentrations of NH4

+, NO2− and NO3

− were measured bya continuous flow analyser (Bran + Luebbe GmbH Inc.,Hamburg, Germany).

PLFA experiments

Phospholipid fatty acids were extracted using a modifiedprocedure as previously described (Brant et al., 2006).Briefly, 2 g of soil for each sample was incubated in an extrac-tion mixture containing methanol, chloroform and phosphatebuffer (2:1:0.8, v/v/v) for 2 h. The lipids were separated intoneutral lipids, glycolipids and phospholipids using silica acidcolumns, then phospholipids were methylated to fatty-acidmethyl esters and analysed using a Sherlock Microbial Iden-tification System (MIDI Inc., Newark, DE, USA). The PLFAsi14:0, i15:0, a15:0, 15:0, i16:0, 16:1w5c, i17:0, a17:0, cy17:0,

17:0, 16:1w9c, 18:1w5c, 18:1w7c and cy19:0 were chosen asbacterial markers (Frostegård and Bååth, 1996); while18:1w9c was chosen to represent fungi (Mikola and Setälä,1999). In addition, 10Me17:0, 10Me18:0 were used asmarkers for actinomycetes (Frostegård et al., 1993).

GeoChip experiments

Deoxyribonucleic acid for GeoChip analysis was extractedfrom 5 g of well mixed soil by a freeze-grinding mechanicallysis method as previously described (Zhou et al., 1996).Deoxyribonucleic acid was purified by electrophoresis withlow melting point agarose gel, followed by extraction withphenol–chloroform prior to butanol precipitation. The finalDNA concentrations were quantified with a PicoGreen kit.Deoxyribonucleic acid quality was measured by a Nanodrop(NanoDrop Technologies Inc., Wilmington, DE, USA), and thevalues of 260/280 nm of 1.8, and 260/230 nm of over 1.5were considered acceptable.

Purified DNA (2 μg) was labelled with Cy-5 fluorescent dyeusing random primers, then purified and dried at 45°C for45 min (SpeedVac, ThermoSavant, Milford, MA, USA), priorto suspension in a hybridization buffer [40% formamide,3 × SSC, 10 μg of unlabelled herring sperm DNA (Promega,Madison, WI) and 0.1% SDS]. Labelled DNA was hybridizedwith GeoChip 3.0 at 42°C for 12 h. After washing with1 × SSC–0.2% SDS and 0.1 × SSC–0.2% SDS for 5 min and0.1 × SSC for 30 s, microarrays were scanned (ProScanArray, Perkin Elmer, Waltham, MA, USA), and the signalintensity for each probe was measured by using IMAGENE

software (6.0 premium version, Biodiscovery, El Segundo,CA, USA) (He et al., 2010).

GeoChip data normalization and analysis

Data normalization and analysis were conducted as pre-viously described (Yang et al., 2013; 2014). Raw data wasdownloaded from the website (http://ieg.ou.edu/microarray/)and analysed according to the following steps: (i) poor-qualityspots flagged as 1 or 3 and with SNR less than 2.0 wereremoved, (ii) after removing bad spots, data normalization ofeach spot was performed by logarithm-transforming eachsignal intensity and then dividing the transformed signalintensity by the mean intensity across all genes on themicroarray, and (iii) probes detected in only one replicatewere eliminated as noise.

STATISTICAL ANALYSIS SYSTEM software (SAS Institute, Cary,NC, USA) was used to examine the differences in environ-mental attributes based on one-way analysis of variance(ANOVA) at all three sites. Using an online pipeline (http://ieg.ou.edu/), permutational multivariate analysis of variance(Adonis) based on the Bray–Curtis index to calculate dis-tance matrices was used to examine the treatment effects onfunctional (GeoChip) compositions of soil microbial commu-nities. Detrended correspondence analysis and hierarchicalcluster analysis were performed with the Vegan package inR v.2.12.2 (R Development Core Team, 2011) to illustrate thedifferences in soil microbial community composition amongsamples. Mantel tests were performed based on Bray–Curtisand Euclidean indexes to test the correlations between soilattributes and soil microbial community composition. Canoni-

8 S. Liu et al.


http://ieg.ou.edu/microarray/

http://ieg.ou.edu/

http://ieg.ou.edu/

cal correspondence analysis was performed using Veganpackage in R v.2.12.2. The selected attributes in canonicalcorrespondence analysis (CCA) modelling were based onVIFs, which stepwise removed redundant attributes, resultingin six environmental attributes (annual average temperature,pH, relative humidity, AN, NH4

+-N and aboveground biomass)with VIFs less than 20. Pearson correlation coefficients werecalculated to determine correlations between functionalgenes and selected soil attributes or soil ecoprocesses. Also,Pearson correlations between soil and climate attributeswere calculated. Significant differences between treatmentsand control were analysed by two-tailed t-tests based ontwo-sample equal variance. The significance of interactiveeffects of soil transplant and maize cropping was tested bytwo-way ANOVA in SPSS 19.0 (SPSS Inc., Chicago, IL, USA).

Acknowledgements

The authors thank Hailun, Fengqiu and Yingtan ResearchStation staff for sampling assistance and the anonymousreviewers for constructive comments and suggestions for theimprovement of this manuscript. This research was sup-ported by grants to Yunfeng Yang from the National NaturalScience Foundation of China (41171201) and State KeyJoint Laboratory of Environment Simulation and PollutionControl (11Z03ESPCT); grants to Bo Sun from National BasicResearch Program of China (2011CB100506) and theNational Natural Science Foundation of China (41271258);and grants to Jizhong Zhou from the United States Depart-ment of Energy, Biological Systems Research on theRole of Microbial Communities in C Cycling Program (DE-SC0004601), the U.S. National Science Foundation underthe contract (NSF EF-1065844) and by the United StatesDepartment of Agriculture (Project 2007–35319-18305)through NSF-USDA Microbial Observatories Program. Thisstudy made use of GeoChip and associated computationalpipelines whose development was funded by ENIGMA –Ecosystems and Networks Integrated with Genes andMolecular Assemblies through the Office of Science, Office ofBiological and Environmental Research, the US Departmentof Energy under Contract No. DE-AC02-05CH11231.

Preliminary data accessibility section

All of the GeoChip data will be accessible from GEO data-base (http://www.ncbi.nlm.nih.gov/geo/). The accessible num-ber is GSE51592.

References

Alley, R.B., Marotzke, J., Nordhaus, W., Overpeck, J., Peteet,D., Pielke, R., et al. (2003) Abrupt climate change. Science299: 2005–2010.

Brant, J.B., Sulzman, E.W., and Myrold, D.D. (2006) Micro-bial community utilization of added carbon substrates inresponse to long-term carbon input manipulation. Soil BiolBiochem 38: 2219–2232.

Breeuwer, A., Heijmans, M.P.D., Robroek, B.M., andBerendse, F. (2010) Field simulation of global change:

transplanting Northern Bog Mesocosms Southward. Eco-systems 13: 712–726.

Bryant, J.A., Lamanna, C., Morlon, H., Kerkhoff, A.J.,Enquist, B.J., and Green, J.L. (2008) Microbes on moun-tainsides: contrasting elevational patterns of bacterialand plant diversity. Proc Natl Acad Sci 105: 11505–11511.

Chèneby, D., Philippot, L., Hartmann, A., Hénault, C., andGermon, J.-C. (2000) 16S rDNA analysis for characteriza-tion of denitrifying bacteria isolated from three agriculturalsoils. FEMS Microbiol Ecol 34: 121–128.

Chu, H., Fierer, N., Lauber, C.L., Caporaso, J., Knight, R.,and Grogan, P. (2010) Soil bacterial diversity in the Arctic isnot fundamentally different from that found in other biomes.Environ Microbiol 12: 2998–3006.

Dandie, C.E., Burton, D.L., Zebarth, B.J., Trevors, J.T., andGoyer, C. (2007) Analysis of denitrification genes and com-parison of nosZ, cnorB and 16S rDNA from culturabledenitrifying bacteria in potato cropping systems. Syst ApplMicrobiol 30: 128–138.

Dandie, C.E., Wertz, S., Leclair, C.L., Goyer, C., Burton, D.L.,Patten, C.L., et al. (2011) Abundance, diversity and func-tional gene expression of denitrifier communities in adja-cent riparian and agricultural zones. FEMS Microbiol Ecol77: 69–82.

De Frenne, P., Brunet, J., Shevtsova, A., Kolb, A., Graae,B.J., Chabrerie, O., et al. (2011) Temperature effects onforest herbs assessed by warming and transplant experi-ments along a latitudinal gradient. Glob Change Biol 17:3240–3253.

Falkowski, P.G., Fenchel, T., and Delong, E.F. (2008) Themicrobial engines that drive Earth’s biogeochemical cycles.Science 320: 1034–1039.

Fierer, N., and Jackson, R.B. (2006) The diversity and bio-geography of soil bacterial communities. Proc Natl AcadSci USA 103: 626–631.

Frey, S.D., Elliott, E.T., and Paustian, K. (1999) Bacterial andfungal abundance and biomass in conventional andno-tillage agroecosystems along two climatic gradients.Soil Biol Biochem 31: 573–585.

Frostegård, A., and Bååth, E. (1996) The use of phospholipidfatty acid analysis to estimate bacterial and fungal biomassin soil. Biol Fertil Soils 22: 59–65.

Frostegård, Å., Bååth, E., and Tunlio, A. (1993) Shifts in thestructure of soil microbial communities in limed forests asrevealed by phospholipid fatty acid analysis. Soil BiolBiochem 25: 723–730.

Fuhrman, J.A., Steele, J.A., Hewson, I., Schwalbach, M.S.,Brown, M.V., Green, J.L., and Brown, J.H. (2008) A latitu-dinal diversity gradient in planktonic marine bacteria. ProcNatl Acad Sci USA 105: 7774–7778.

Gao, H., Yang, Z.K., Barua, S., Reed, S.B., Romine, M.F.,Nealson, K.H., et al. (2009) Reduction of nitrate inShewanella oneidensis depends on atypical NAP and NRFsystems with NapB as a preferred electron transportprotein from CymA to NapA. ISME J 3: 966–976.

García-Orenes, F., Guerrero, C., Roldán, A., Mataix-Solera,J., Cerdà, A., Campoy, M., et al. (2010) Soil microbialbiomass and activity under different agricultural manage-ment systems in a semiarid Mediterranean agroe-cosystem. Soil Tillage Res 109: 110–115.



http://www.ncbi.nlm.nih.gov/geo/

García-Ruiz, R., Ochoa, V., Hinojosa, M.B., and Carreira, J.A.(2008) Suitability of enzyme activities for the monitoring ofsoil quality improvement in organic agricultural systems.Soil Biol Biochem 40: 2137–2145.

Hansen, J., Sato, M., Ruedy, R., Lo, K., Lea, D.W., andMedina-Elizalde, M. (2006) Global temperature change.PNAS 103: 14288–14293.

Hare, S.R., and Mantua, N.J. (2000) Empirical evidence forNorth Pacific regime shifts in 1977 and 1989. ProgOceanogr 47: 103–145.

He, Z., Deng, Y., Van Nostrand, J.D., Tu, Q., Xu, M., Hemme,C.L., et al. (2010) GeoChip 3.0 as a high-throughput toolfor analyzing microbial community composition, structureand functional activity. ISME J 4: 1167–1179.

Hirsch, P.R., Mauchline, T.H., and Clark, I.M. (2010) Culture-independent molecular techniques for soil microbialecology. Soil Biol Biochem 42: 878–887.

Jung, J., Yeom, J., Kim, J., Han, J., Lim, H.S., Park, H., et al.(2011) Change in gene abundance in the nitrogenbiogeochemical cycle with temperature and nitrogen addi-tion in Antarctic soils. Res Microbiol 162: 1018–1026.

Khan, S.A., Mulvaney, R.L., and Hoeft, R.G. (2001) A simplesoil test for detecting sites that are nonresponsive tonitrogen fertilization this study was a part of Project ILLU-65-0326, Illinois Agric. Exp. Stn. Soil Sci Soc Am J 65:1751–1760.

King, G.M., and Adamsen, A.P.S. (1992) Effects of tempera-ture on methane consumption in a forest soil and in purecultures of the Methanotroph Methylomonas rubra. ApplEnviron Microbiol 58: 2758–2763.

Lagomarsino, A., Moscatelli, M.C., Di Tizio, A., Mancinelli, R.,Grego, S., and Marinari, S. (2009) Soil biochemical indica-tors as a tool to assess the short-term impact of agriculturalmanagement on changes in organic C in a Mediterraneanenvironment. Ecol Indic 9: 518–527.

Lauber, C.L., Hamady, M., Knight, R., and Fierer, N. (2009)Pyrosequencing-based assessment of soil pH as a predic-tor of soil bacterial community structure at the continentalscale. Appl Environ Microbiol 75: 5111–5120.

Lejon, D.P.H., Nowak, V., Bouko, S., Pascault, N., Mougel,C., Martins, J.M.F., and Ranjard, L. (2007) Fingerprintingand diversity of bacterial copA genes in response to soiltypes, soil organic status and copper contamination. FEMSMicrobiol Ecol 61: 424–437.

Li, Y.-T., Rouland, C., Benedetti, M., Li, F., Pando, A., Lavelle,P., and Dai, J. (2009) Microbial biomass, enzyme andmineralization activity in relation to soil organic C, N and Pturnover influenced by acid metal stress. Soil Biol Biochem41: 969–977.

Liang, Y., Van Nostrand, J.D., N’Guessan, L.A., Peacock,A.D., Deng, Y., Long, P.E., et al. (2012) Microbial functionalgene diversity with a shift of subsurface redox conditionsduring in situ uranium reduction. Appl Environ Microbiol 78:2966–2972.

Ma, X., Chen, T., Zhang, G., and Wang, R. (2004) Microbialcommunity structure along an altitude gradient in threedifferent localities. Folia Microbiol 49: 105–111.

McAnena, A., Flögel, S., Hofmann, P., Herrle, J., Griesand,A., Pross, J., et al. (2013) Atlantic cooling associated witha marine biotic crisis during the mid-Cretaceous period.Nat Geosci 6: 558–561.

Mebius, L.J. (1960) A rapid method for the determination oforganic carbon in soil. Anal Chim Acta 22: 120–124.

Mettel, C., Kim, Y., Shrestha, P.M., and Liesack, W. (2010)Extraction of mRNA from soil. Appl Environ Microbiol 76:5995–6000.

Mi, L., Wang, G., Jin, J., Sui, Y., Liu, J., and Liu, X. (2012)Comparison of microbial community structures in fourBlack soils along a climatic gradient in northeast China.Can J Soil Sci 92: 543–549.

Mikola, J., and Setälä, H. (1999) Interplay of omnivory,energy channels and C availability in a microbial-based soilfood web. Biol Fertil Soils 28: 212–218.

Montalvo, A.M., and Ellstrand, N.C. (2001) Transplantation ofthe subshrub lotus scoparius: testing the home-site advan-tage hypothesis. Conserv Biol 14: 1034–1045.

Morales, S.E., Cosart, T., and Holben, W.E. (2010) Bacterialgene abundances as indicators of greenhouse gas emis-sion in soils. ISME J 4: 799–808.

Murphy, J., and Riley, J.P. (1962) A modified single solutionmethod for the determination of phosphate in naturalwaters. Anal Chim Acta 27: 31–36.

Nevarez, L., Vasseur, V., Le Madec, A., Le Bras, M., Coroller,L., Leguérinel, I., and Barbier, G. (2009) Physiological traitsof Penicillium glabrum strain LCP 08.5568, a filamentousfungus isolated from bottled aromatised mineral water. IntJ Food Microbiol 130: 166–171.

Nilsson, L.O., Bååth, E., Falkengren-Grerup, U., andWallander, H. (2007) Growth of ectomycorrhizal myceliaand composition of soil microbial communities in oak forestsoils along a nitrogen deposition gradient. Oecologia 153:375–384.

Olsen, S.R., Cole, C., and Watanabe, F.S. (1954) Estimationof available phosphorus in soils by extraction with sodiumbicarbonate. United States Department of Agriculture Cir-cular No. 939. Banderis, A. D., Barter, D. H., and Anderson,K. Agricultural and Advisor.

Page, A.L. (1982) Methods of Soil Analysis. Part 2. Chemicaland Microbiological Properties: American Society of Agro-nomy. Madison, WI, USA: Soil Science Society of America.

Peeters, K., and Willems, A. (2011) The gyrB gene is a usefulphylogenetic marker for exploring the diversity ofFlavobacterium strains isolated from terrestrial and aquatichabitats in Antarctica. FEMS Microbiol Lett 321: 130–140.

Petchey, O.L., McPhearson, P.T., Casey, T.M., and Morin,P.J. (1999) Environmental warming alters food-web struc-ture and ecosystem function. Nature 402: 69–72.

Potthoff, M., Steenwerth, K.L., Jackson, L.E., Drenovsky,R.E., Scow, K.M., and Joergensen, R.G. (2006) Soil micro-bial community composition as affected by restorationpractices in California grassland. Soil Biol Biochem 38:1851–1860.

R Development Core Team (2011) R: A Language andEnvironmental for Statistical Computing. Vienna, Austria:R Foundation for statistical computing.

Reed, H.E., and Martiny, J.B.H. (2007) Testing the functionalsignificance of microbial composition in natural communi-ties. FEMS Microbiol Ecol 62: 161–170.

Rinnan, R., Michelsen, A., Bååth, E., and Jonasson, S.(2007) Fifteen years of climate change manipulations altersoil microbial communities in a subarctic heath ecosystem.Glob Change Biol 13: 28–39.

10 S. Liu et al.


Roesch, L.F.W., Fulthorpe, R.R., Riva, A., Casella, G.,Hadwin, A.K.M., Kent, A.D., et al. (2007) Pyrosequencingenumerates and contrasts soil microbial diversity. ISME J1: 283–290.

Rousk, J., Bååth, E., Brookes, P.C., Lauber, C.L., Lozupone,C., Caporaso, J.G., et al. (2010) Soil bacterial and fungalcommunities across a pH gradient in an arable soil. ISMEJ 4: 1340–1351.

Rowe, E.C., Emmett, B.A., Frogbrook, Z.L., Robinson, D.A.,and Hughes, S. (2012) Nitrogen deposition and climateeffects on soil nitrogen availability: influences of habitattype and soil characteristics. Sci Total Environ 434: 62–70.

Savolainen, O., Pyhäjärvi, T., and Knürr, T. (2007) Gene flowand local adaptation in trees. Annu Rev Ecol Evol Syst 38:595–619.

Sessitsch, A., Gyamfi, S., Stralis-Pavese, N., Weilharter, A.,and Pfeifer, U. (2002) RNA isolation from soil for bacterialcommunity and functional analysis: evaluation of differentextraction and soil conservation protocols. J MicrobiolMethods 51: 171–179.

Sharma, S.K., Ramesh, A., Sharma, M.P., Joshi, O.P.,Govaerts, B., Steenwerth, K.L., and Karlen, D.L. (2011)Microbial community structure and diversity as indi-cators for evaluating soil quality. In Biodiversity, Biofuels,Agroforestry and Conservation Agriculture. Lichtfouse, E.(ed.). Netherlands: Springer, pp. 317–358.

Shaw, M.R., and Harte, J. (2001) Response of nitrogencycling to simulated climate change: differential responsesalong a subalpine ecotone. Glob Change Biol 7: 193–210.

Van Nostrand, J.D., Wu, L., Wu, W.-M., Huang, Z., Gentry,T.J., Deng, Y., et al. (2011) Dynamics of microbial commu-nity composition and function during in situ bioremediationof a uranium-contaminated aquifer. Appl Environ Microbiol77: 3860–3869.

Vanhala, P., Karhu, K., Tuomi, M., Björklöf, K., Fritze, H.,HyvÄRinen, H., and Liski, J. (2011) Transplantation oforganic surface horizons of boreal soils into warmerregions alters microbiology but not the temperature sensi-tivity of decomposition. Glob Change Biol 17: 538–550.

Walther, G.R. (2010) Community and ecosystem responsesto recent climate change. Philos Trans R Soc B Biol Sci365: 2019–2024.

Wardle, D.A., Bardgett, R.D., Klironomos, J.N., Setala, H.,van der Putten, W.H., and Wall, D.H. (2004) Ecologicallinkages between aboveground and belowground biota.Science 304: 1629–1633.

Yang, Y., Wu, L., Lin, Q., Yuan, M., Xu, D., Yu, H., et al.(2013) Responses of the functional structure of soil micro-bial community to livestock grazing in the Tibetan alpinegrassland. Glob Change Biol 19: 637–648.

Yang, Y., Gao, Y., Wang, S., Xu, D., Yu, H., Wu, L., et al.(2014) The microbial gene diversity along an elevationgradient of the Tibetan grassland. ISME J 8: 430–440.

Yergeau, E., Kang, S., He, Z., Zhou, J., and Kowalchuk, G.A.(2007a) Functional microarray analysis of nitrogen andcarbon cycling genes across an Antarctic latitudinaltransect. ISME J 1: 163–179.

Yergeau, E., Bokhorst, S., Huiskes, A.H.L., Boschker, H.T.S.,Aerts, R., and Kowalchuk, G.A. (2007b) Size and structureof bacterial, fungal and nematode communities along an

Antarctic environmental gradient. FEMS Microbiol Ecol 59:436–451.

Zhang, L.-M., Hu, H.-W., Shen, J.-P., and He, J.-Z. (2011)Ammonia-oxidizing archaea have more important role thanammonia-oxidizing bacteria in ammonia oxidation ofstrongly acidic soils. ISME J 6: 1032–1045.

Zhang, W., Parker, K.M., Luo, Y., Wan, S., Wallace, L.L., andHu, S. (2005) Soil microbial responses to experimentalwarming and clipping in a tallgrass prairie. Glob ChangeBiol 11: 266–277.

Zhou, J., Bruns, M.A., and Tiedje, J.M. (1996) DNA recoveryfrom soils of diverse composition. Appl Environ Microbiol62: 316–322.

Zhou, J., Xue, K., Xie, J., Deng, Y., Wu, L., Cheng, X., et al.(2012) Microbial mediation of carbon-cycle feedbacks toclimate warming. Nat Clim Chang 2: 106–110.

Supporting information

Additional Supporting Information may be found in the onlineversion of this article at the publisher’s web-site:

Fig. S1. The effect of soil transplant on microbial alpha-diversity in bare fallow samples. Alpha-diversity was calcu-lated by Simpson index. S: soil samples in southern China(Yintan); SC: soil samples transplanted from southern Chinato central China (Fengqiu); SN: soil samples transplantedfrom southern China to northern China (Hailun). Data arepresented as mean ± standard errors. Significant differencesbetween S and SC or between S and SN were analysed bytwo-tailed t-tests with two-sample equal variance and indi-cated by ‘*’ for P < 0.1 and ‘**’ for P < 0.05 respectively.Fig. S2. Comparison of microbial group biomass in PLFAexperiments for bare fallow samples. Data are presented asmean ± standard errors. Significant differences between Sand SC or between S and SN were analysed by two-tailedt-tests with two-sample equal variance and indicated by ‘**’for P < 0.05. Symbols as in Supporting information Fig. S1.Fig. S3. Detrended correspondence analysis of soil micro-bial communities for all of soil samples based on GeoChipdata. Symbols as in Supporting information Fig. S1; m: maizecropping.Fig. S4. Observed versus additive effects of soil transplantand maize cropping on microbial alpha-diversities. Theobserved effect = SCm (or SNm) – S. The additive effect= SC (or SN) – S + Sm – S. Significant interactive effectswere indicated by ‘***’ for P < 0.01. Symbols as in Supportinginformation Fig. S3.Table S1. Comparison of gene category distribution at differ-ent sites.Table S2. Genes detected only at site Sa, and at neither sitesSC nor SN.Table S3. The statistical test of Adonisa to analyse theeffects of dual factors of soil transplant and maize croppingon soil microbial community compositions measured byGeoChip.Table S4. Environmental attributes of samples with maizecropping.Table S5. Genes significantly (P < 0.01) correlated withannual average temperature.Table S6. Genes significantly correlated (P < 0.01) with pH.



the interactive effects of soil transplant into colder regions and cropping on soil microbiology and...

Documents